Hello, and welcome back. My name is Tyler McMinn with Aruba at Hewlett Packard Enterprise Company, and this is the wired vs wireless video in our Aruba Mobility Essentials course, our second video in our part one series on wireless technology. Without further ado, let's get started. Wired vs wireless, what is there to say? Well, let's take a comparison between the wired on the left-hand side, and wireless on the right-hand side. What are some of the key differences here? On the wire when you're transmitting your data from one end of a cable, like this ethernet cable, to the other, the voltage and the changes that represent one and zero, are going to pass along this cable. It's corralled into the cable. When it comes to a wireless radio like this Aruba hospitality access point that I've had for a while here. The radio inside will transmit out an antenna, that's also happens to be internal on this guy, and broadcasts it's signal out so that it's much more easy to connect using a mobile device such as your phone or a laptop or whatever you'd like. On the wired side, what we can add to this, is that wired transmits data bounded to the wires themselves, whether it's electrical impulses of changes of voltage to represent ones and zeros, or it's little beams of light that indicate photons and then indicate whether it's a one or a zero. It is typically a fixed location so you know exactly where the other end of that cable is going to be punched into a wall or into a docking station or something like that, and it is also contention based. Now, what does contention based mean? Well, this refers to the old half-duplex communication that we would have with our hubs. If hubs sounds familiar, it should. These had been replaced by switches today to where we don't have the same issues that hubs had, hubs transmitted everything on a shared medium. What that means is that if user A, or B, or C here went to send a frame or a bit of information, that data would utilize the entire hub and be blasted out all ports that were connected on that hub, even back the way that it came on a receiving wire. Why this half-duplex Carrier Sense Multiple Access Collision Detection feature? Well, this feature was a way that your computers here, your hosts, station A, station B, and station C, could all contend with sharing the same hub. If only one person could send at the same time, otherwise you'd have a collision, then you would need some mechanism to detect collisions and to avoid them should they occur. What that means is if A were to send a frame at the same time that B also happened to be listening and had backed off and had now decided no one was sending, and so both A and B, both jump on the hub at the same time, you're going to get an overlap of bits indicating a collision to all the recipients, meaning that A and B both realize that their data is not going to be understood. The way that they were resolved that, is they would send a jam signal to all the other devices on the hub, indicating that everybody should just back off some random amount of milliseconds, listen for the line once they're timer ran out, and then if the line was clear, go ahead and send. If not, if someone was sending already, they'd back off again a random amount of time, then count down until it was there time to send. This mechanism work great for years and years and years through the use of hubs. With switches, it's a little bit different in that switches every cable, every device that's connected is its own little collision domain. Where as a hub you share a collision domain with all the devices plugged into the hub, with the switch, each wire is its own little collision domain, such that you can't really have a collision since you have transmission wires and wires that receive on the same cable, there's no possibility of a collision unless there's some other issue going on. With switches these days, we generally enjoy a zero collision environment, and therefore can run at what we call full-duplex communication. This whole discussion was something that we covered in a previous set of videos available to you guys known as the switching essentials or switching basics course. If you want to check that course out, we do a huge deep dive into the switching part of it. But suffice to say that's enough for us to know what's going on on the wireless side, because on the wireless side, APs behave much more like a hub, then they do a switch. What I mean by that is they share the medium. They have contention based access for their wireless users, the same way that a hub shares the medium with it's wired users. Users A, B, and C over here are wirelessly connected, but they're all connected on the same channel, for example, say channel six and the 2.4 gigahertz. What does that mean? That means if A goes to send, and B goes to send at the same time, they have a collision, and so we need some mechanism to deal with that as well. Whereas on the wired side with switches. Switches, you don't have collisions and therefore you can run at full-duplex. This gives wired a bit of an advantage and that aspect over wireless, where on hubs it is true only two devices on a cable, client and switch and ultimately on the wired side, you would have a better or faster experience when it came to collisions because you could avoid them. On the wireless portion, I'll clear this out. On the wireless comparatively, they transmit using radio frequency which is unbounded and this is the key advantage of wireless over wired. Sure, wired is considered faster, and so we use wired in scenarios with switches such as, if you're connecting servers or in a data center or a high application devices like servers or printers or video, like a video streaming recorder or player or something like that, it's recommended to move those off the wireless in most cases, or if you're plugging in an access point. Access points themselves are trying to bridge dozens of users at the same time all on a single cable, so they eat up that bandwidth pretty quickly using one gigabit per second or 2.5 gigabytes, or even five gigs, even 10 gigabits per second depending on the use case there. We need switches that can not only handle that but it would be almost a nightmare to try and do that wirelessly. You could wirelessly repeat a signal to an access point that could get power but not wired connectivity. Generally though, we would prefer to wire all of our access points. Wireless transmits using RF on an unbounded medium, meaning that you have the flexibility of connecting without a cable. That's awesome. Obviously, as soon as people had that technology in their hands, since then, most people have moved to predominantly wireless environments unless they had some special use cases like the ones I described there. They wire up their access point and then they bridge their wireless devices. For a lot of us, the technology's at the point now, we don't even think about it. We make phone calls on wireless, we do printing on wireless, we do our secure banking on wireless, wireless is predominantly the main way that most people will use to connect their general devices. This mobility as long as it's within the RF area, is huge. It's a huge advantage over running cables everywhere, and therefore it's taken over. It is however not all good news, it is contention based, so you need some collision detection mechanism for the reasons I described, an access point acts much more like a hub in that way than a switch and so we do the same thing. We have carrier sense where you can tell who's connected, the multiple access, availability of an access point and then instead of collision detection where everybody on a hub knows who's connected. With access points, it's a little more complicated and that you might have devices or users that are so far away from the AP, they can still connect but they are not being seen by devices on the other side of that access point. We have this idea of collision avoidance where we can actually register with an access point to determine is it safe to send or not, otherwise, you deal with collisions. Collisions on the same channel where you have multiple APs and users are bouncing against each other or they're trying to utilize the same Channel 6, or your neighbor is trying to use the same Channel 6, this is known as Co-channel interference, CCI. We're going to have a deeper discussion on that in a few videos here but for now, this is easily one of the biggest issues that we run into when designing proper wireless. It's an issue we need to overcome. With this shared medium, you ultimately result in a slower experience generally than wireless. Over the years, we'll get into this more with a different IEEE standards. We've seen our speeds go from one, maybe two megabits per second with the original 802.11 to around 10 megabits per second average with g and a speeds theoretically 54 megabits but in reality, more closer to 10 to hundreds of megabits per second with 802.11 in and now with the newest high-throughput standards like 802.11AC and AX. We're seeing for the first time in gigabits per second of throughput which really makes wireless at least at the moment, very competitive as an offering compared to wired. Assuming you can get all the little issues cleared up and that's what we're going to discuss in this series of videos. I hope that this discussion comparing wired vs wireless really helped to explain some of the major differences between the two and some of the issues that both share, whereas some are going to be only wireless issues versus wired. In a modern network, we typically don't have these contention based issues like we used to whereas on the wireless side, this is one of the major hurdles that go into how we design wireless and how we fix it. Again, thank you for your time. I appreciate it. I'm going to see you guys in the next video where we're going to go over wireless LAN organizations. I'll see you then.
